Autonomous IoT system for kinematic monitoring of wrist rehabilitation

Abstract

This work addresses the pressing scientific and practical challenge of objectifying the process of medical telemetric rehabilitation. To this end, an innovative autonomous IoT system is proposed for the objective, real-time kinematic monitoring of wrist rehabilitation. The system is based on an autonomous mobile device that employs an edge computing paradigm to provide objective, quantitative monitoring of wrist motor activity. Key implementation requirements were established: the collection of telemetric data—encompassing not only statistical wrist movement parameters but also kinematic indicators (such as physiological and neurological tremor)—and the generation of interactive visual feedback to maintain patient cognitive engagement. The hardware architecture, mathematical modeling, and software implementation for this autonomous mobile device were developed. The device utilizes a high-performance microcontroller and an inertial measurement unit equipped with an integrated digital motion processor. The selection of the microcontroller and necessary sensors is justified. Particular attention is paid to the "analytical engine," which calculates the wrist tilt angle, angular velocity magnitude, and an integral smoothness score in real time. The algorithmic implementation of this engine leverages the powerful mathematical framework of hypercomplex numbers and resolves the "gimbal lock" issue when calculating spatial orientation. To ensure an accurate count of movements amidst physiological tremor, an adaptive detector based on a finite state machine with hysteresis was implemented. To guarantee maximum data reliability, the results of each session are stored locally in a relational database. To enhance patient engagement in the rehabilitation process, the implementation of a robust module featuring therapeutic mini-games has been proposed. A key advantage of the developed autonomous mobile device architecture is its independence from cloud services, ensuring the confidentiality of medical data. A web interface has been implemented to ensure low latency during the visualization of movement trajectories. Future work will focus on field-testing this autonomous IoT system for real-time, objective kinematic monitoring during wrist rehabilitation in clinical settings.